HJBR Sep/Oct 2023

HEALTHCARE JOURNAL OF BATON ROUGE I  SEP / OCT 2023 57 Sotirios Stathakis, PhD, DABR, FAAPM Chief of Physics Mary Bird Perkins Cancer Center AI organ segmentation can lead to a higher patient throughput and higher quality treat- ment plans, hence better patient outcomes. TUMOR DETECTION Another application of AI is used to iden- tify tumors in the brain. For example, a pro- cess combining advanced imaging technol- ogy and AI can accurately diagnose brain tumors in fewer than three minutes. This approach is able to accurately distinguish tumor tissue from healthy tissue. Artificial intelligence has shown great potential in improving the accuracy and efficiency of brain tumor identification for stereotactic radiosurgery (SRS). With AI assistance, the inter-reader agreement can significantly in- crease, and algorithm-assisted physicians demonstrate a higher sensitivity for lesion detection than unassisted physicians. AI as- sistance can improve contouring accuracy, especially for physicians with less SRS expe- rience, and can also improve efficiency. Sev- eral studies show results that suggest deep learning neural networks can be optimally utilized to improve accuracy and efficiency for the clinical workflow in brain tumor SRS. SYNTHETIC IMAGE GENERATION One of the latest applications of AI in ra- diation therapy is the creation of synthetic image sets. The most common example is the creation of a synthetic CT (sCT) from MRI images. MRI is a powerful imaging mo- dality used in radiation therapy for tumor visualization and treatment planning. How- ever, traditional treatment planning requires CT images due to its better tissue density information for dose calculations. MRI sCT application involves the generation of syn- thetic CT images from MRI data, allowing treatment planning to be performed directly on MRI scans. Advantages of MRI sCT ap- plication include: (a) elimination of CT Im- aging —MRI sCT application eliminates the need for separate CT imaging, simplifying the treatment planning process and reduc- ing the burden on patients; (b) superior soft tissue visualization; (c) enhanced accuracy as the synthetic images more accurately represent the patient’s anatomy compared to using CT images alone; and (d) uses in adaptive radiation therapy when combined with MRI linear accelerators for treatment delivery. In conclusion, AI has shown great po- tential in improving the accuracy and effi- ciency of radiation therapy. With the help of AI, physicians can achieve higher sensitivity for lesion detection, improved contouring accuracy, and increased inter-reader agree- ment. Additionally, AI can significantly re- duce the time required for the clinical work- flow, allowing for more efficient treatment and better quality treatment plans. Overall, the integration of AI into radiation therapy has the potential to greatly improve patient outcomes. n Sotirios Stathakis, PhD, DABR, FAAPM, serves as Mary Bird Perkins Dr.CharlesM.Smith chief of phys- ics. In this role,he oversees the overall management and oversight of the cancer center’s physics and do- simetry teams, in support of clinical, research, and educational activities.Stathakis obtained a Bachelor of Science in honors physics from the University of Waterloo in Canada, a Master of Science in medi- cal physics from the Medical Physics and Bio-Engi- neering Department at the University of Aberdeen, Scotland, UK, and a PhD in medical physics from the University of Patras, Hellas, Greece. RADIATION THERAPY is a type of cancer treatment that uses high-energy radiation to kill cancer cells. It works by damaging the DNA of cancer cells, which prevents them frommultiplying and growing. This therapy is typically used in combination with other cancer treatments, such as chemotherapy and surgery. State of the art technological advances in radiation therapy delivery systems and in computer sciences are available and utilized to precisely and accurately treat patients. The latest of these is the application of arti- ficial intelligence (AI) as a cancer treatment. Some of the applications of AI include auto- mated organ segmentation, tumor detection, and synthetic image generation. ORGAN SEGMENTATION AI is being used in organ segmentation for radiation therapy to improve the effi- ciency and accuracy of the process. Organ segmentation is a crucial, labor-intensive step in radiation oncology that can often turn into a clinical workflow bottleneck. It involves identifying the organs and tissues in diagnostic images that must be targeted or protected during radiation therapy and can take hours per patient. AI models for organ segmentation are being developed to automate this process andmake it faster and more accurate. Studies show that AI-based contouring significantly reduces the time required for organ segmentation compared tomanual contouring and has a high level of accuracy. AI-based contouring can also re- duce inter-observer variability in organ seg- mentation and improve the consistency of target volumes among radiation oncologists. The speed, accuracy, and standardization of